Fuzzy mathematics based fusion method for accident reconstruction results

The fusion of accident reconstruction results is crucial due to their inherent uncertainty. When there are potential risk results, there are few methods to deal with such cases. The characteristics of fuzzy mathematics are highly consistent with the uncertainty of accident reconstruction, so this paper proposed a fuzzy mathematics-based accident reconstruction results fusion method. This method first determines the membership function corresponding to each model result based on the obtained model reconstruction results and subintervals. Based on the membership functions, the relationship between subinterval and model result is transformed into fuzzy set. Afterward, it calculates the value of the evaluation index of the fuzzy set. When the obtained value is low, consider that the model reconstruction result corresponding to this value may be a potential risk result, and its probability proportion needs to be reduced. Then combine it with the fuzzy set operation to obtain the final fuzzy set expression. Finally, it normalizes the membership degree in the fuzzy set expression, and the resulting can be considered as the probability of the subintervals of the final fusion results. Both numerical and real case results show that the proposed method can ensure that the final fused results follow a normal distribution in the process of fusing accident reconstruction results. A further analysis demonstrates that the presence of more than two asymmetric membership functions causes the final fusion results to not follow a normal distribution. The research and analysis provide support for reaching more objective accident reconstruction results.